Rufus Catchings on Pinning Down California’s Faults
Listen to the podcast here or wherever you get your podcasts.
California is riddled with faults. Although some of the most active faults are revealed by surface features, the majority are invisible unless probed by geophysical methods. Fortunately, faults affect the speed and amplitude of seismic waves. As Rufus Catchings explains in the podcast, we can detect such speed and amplitude changes when seismic waves propagate through the ground. Using an array of seismometers to measure seismic wave speeds, we can then generate an image of subsurface seismic speeds and see exactly where the faults lie.
Rufus Catchings is a Research Geophysicist at the US Geological Survey (USGS). Over the past 40 years, he has studied many dozens of faults in California and elsewhere to pin down their precise locations and help assess the risks they pose.
In the image, Catchings is installing a seismometer on the roof of a building in Turkey near the devastating 6 February 2023 Mw 7.8 Turkey Earthquake.
Podcast Illustrations
Images courtesy of Rufus Catchings unless otherwise indicated.
Map of the major faults in California. The thick red trace marks the San Andreas Fault. In the podcast, Catchings discusses faults in the vicinity of San Francisco and Los Angeles, as well as others elsewhere in the US and internationally.
Courtesy of the California Department of Conservation, California Geological Survey
USGS Seismic Imaging Group in the Field
Placing a linear array of seismometers on the water’s edge of the San Francisco Bay.
Conducting a seismic survey in the City of Vallejo, near Napa, California. A 500-lb accelerated weight drop is used as the seismic energy source, and seismic sensors (under the cones) record the resulting seismic energy.
Conducting a seismic survey across the Serra Fault in the City of San Carlos on the San Francisco Peninsula. A 1100-lb weight (attached to the skid steer) is used to generate the seismic energy source, and the seismic energy is recorded with a long array of seismometers, one of which is below the cone.
Preparing for a seismic survey across the San Andreas Fault in Central California. The third person from the camera is standing on the active main trace of the San Andreas Fault at the base of the hill. The white devices are seismometers that are placed in the boreholes. Boreholes with red flags show locations of deeper holes in which explosives are detonated to generate seismic energy. The seismic profile extends to the next hill so that all nearby traces of the San Andreas Fault can be imaged.
Seismic gun source (Catchings)
Catchings uses a seisgun to generate seismic energy used to image the San Andreas Fault in Central California. A seisgun shoots 400-grain black-powder shotgun blanks into the ground surface.
USGS and Turkish scientists deploying a linear array of seismometers in the city of Gaziantep, near the epicenter of the 6 February 2023 Mw 7.8 Turkish earthquake epicenter.
Main Types of Fault
The San Andreas Fault, which defines the boundary between the North American Plate and the Pacific Plate, is a strike-slip fault. As Catchings says in the podcast though, movement along reverse faults can often produce the most violent shaking for a given size earthquake.
Trista L. Thornberry-Ehrlich (Colorado State University)
Guided Waves
Fault zones can act as waveguides and trap seismic waves. Although the fault zone is a region where seismic waves travel more slowly, the wave-guiding effect causes the amplitude of seismic waves to increase. In the podcast, Catchings describes studies in which he used the peak seismic wave amplitudes, measured as peak ground velocity, i.e., shaking movement, to indicate the surface trace of a fault.
Guided waves are transmitted only along continuously connected faults. In the diagrams at right, seismic energy enters a fault zone from the bottom. The dark lines mark the surface, and seismic traces are plotted above the dark line. In the top row, the effects of various subsurface-connected fault configurations on the amplitude of guided waves are shown. In the bottom row in the left example, the faults are only slightly connected at the edge, and some guided wave energy is transmitted. In the other two bottom-row examples, the faults are not connected, and the seismic traces show that no guided waves are transmitted.
After Li, Y. G. & Vidale, J. E. (1996), Bulletin of the Seismological Society of America, 86, 371
This example illustrates the use of peak ground velocity (PGV) to locate the surface trace of a fault. The plot at top shows the peak ground velocity, i.e., amplitude of a seismic wave, along a line from west to east. The peak is located where the fault intersects the surface. Seismic wave speeds are indicated — 3 km/s outside the fault zone and 2 km/s within the fault zone.
Guided Wave Study Across the San Andreas Fault 1906 Rupture Zone
Catchings, R. D., et al. (2020), Bulletin of the Seismological Society of America, 110, 3088
Left: Near San Francisco, the San Andreas Fault runs through San Andreas Lake. The San Andreas Fault was named after this lake following the 1906 San Francisco earthquake. Center: Catchings and his colleagues generated an explosion on the active trace of the San Andreas Fault (red blot at top) and measured seismic wave speeds and ground velocity amplitudes along the seismic profile line marked on the map. The red lines show where the surface ruptured in the 1906 earthquake (mapped by Schussler, 1906), and the red arrows indicate where the rupture lines intersect the profile. The results of the study are shown below. Right: Google Earth image showing the San Francisco Peninsula and Bay. The white box shows the location of the study area along the San Francisco Peninsula.
Fault Locations from Peak Ground Velocities and S-Wave Velocities
Top: Peak seismic wave amplitudes of guided S waves along the profile shown in the previous figure. The known traces of the San Andreas Fault correspond to high peak ground velocities, confirming the technique as a way to identify surface fault locations. Bottom: Tomographic image of S-wave velocities along the seismic profile. A low-velocity region marks two of the 1906 rupture traces.
Using the 2014 Napa Earthquake Guided Waves to Infer Continuity Between Faults
Catchings, R.D., et al. (2016), Bulletin of the Seismological Society of America, 106, 2721, doi: 10.1785/0120160154
Google Earth image of the northeastern San Francisco Bay area showing the 2014 M 6.0 Napa earthquake surface rupture (thick red line); other thin red lines show historically active faults and yellow lines show Quaternary (not historical) faults.
As described in the podcast, Catchings and his team were able to infer that faults to the south (e.g., Franklin Fault) and north of the 2014 surface rupture of the West Napa Fault were connected because seismic guided-wave energy from aftershocks propagated to these faults. The thick red line marks part of the single inferred fault. This increased the inferred fault length tenfold, suggesting that a much larger earthquake is possible on the fault system. The maximum magnitude of an earthquake on a given fault is proportional to the length of the fault. The longer fault also extends to the Maacama Fault to the north and to the Calaveras Fault to the south, suggesting the possibility of an even longer fault system.
Southern California Faults
The map shows some of the more significant faults in southern California. The San Andreas Fault Zone runs diagonally across the middle of the map. The Hollywood fault, which was one of the faults studied by Catchings and discussed in the podcast, runs roughly east-west to the northeast of the dot marking Los Angeles.
Miracosta College
Imaging the Hollywood Fault
Catchings, R.D., et al. (2018), U.S. Geological Survey-California Geological Survey Fault-Imaging Surveys Across the Hollywood and Santa Monica Faults, Los Angeles County, California, Open-File Report 2020-1049 https://doi.org/10.3133/ofr20201049.
In the podcast, Catchings describes imaging the Hollywood Fault by using repetitive artificial impacts to allow subtraction of the “cultural noise” of a dense urban neighborhood.
Line of seismic sensors used to image the Hollywood Fault. The seismic survey was conducted at night to minimize cultural noise.
Seismic-imaging crew planning a night seismic survey in Hollywood, California.
Seismic array recording guided waves at night in Hollywood. Seismometers are white-colored devices between the cones. Guided waves are being generated several blocks away.
Google Earth image of one of the areas studied by Catchings, showing the locations of traces of the Hollywood Fault (yellow lines) and other faults (green lines). The small red stars mark “shot points,” where an accelerated weight or a hammer was used to strike a steel plate on the ground and input seismic energy into the ground. The right blue line (HW1) marks the plane of the seismic wave speed sections shown below.
Tomographic P-wave velocity model along the section HW1, showing the location of nearby streets. Slightly south of the surface trace of the Hollywood Fault, there is a zone of high velocities, especially for velocities above about 1,500 m/s. As mentioned in the podcast, faults can act as groundwater barriers, and the figure is consistent with this interpretation since it suggests that the depth of the top of the water table (1,500 m/s contour) is higher to the south of the fault than to the north of the fault.
As Catchings mentions in the podcast, a high ratio of P-wave to S-wave speeds (Vp/Vs) often provides the clearest indication of the presence of a fault as the broken-up nature of the rocks in a fault zone slows both P- and S-waves down, but the presence of water can mitigate or reverse this effect for P-waves. The profile at right shows a zone of high Vp/Vs near the surface trace of the Hollywood Fault (dashed lines).
S-wave velocities (Vs) along the same section as above, showing a zone of low Vs to a depth of at least 40m. This is consistent with a near-vertical fault.
New Madrid Seismic Zone
The New Madrid Seismic Zone lies near the middle of the North American plate. In 1999, Catchings and his colleagues performed a study of the crustal seismic wave speeds in this region. The results revealed the presence of a thick, dense lower crust, covered by sediment-filled basins. These features are consistent with continental rifting in which the New Madrid Seismic Zone is the failed third arm of the continental rifting whose rifted arms resulted in the Gulf of Mexico.
The map shows earthquakes greater than magnitude 2.5. Red circles: post-1972 earthquakes; blue circles: pre-1973 earthquakes.
geology.com
Induced Seismicity
Catchings, R.D., et al. (2105), J. Geophysical Research Solid Earth, 120, 3479
In the podcast, Catchings described a case of induced seismicity when the increased pore pressure and weight from a reservoir caused a damaging M 6.3 earthquake in 1967 in India. Top: Map showing the location of the Koyna and Warna reservoirs. Bottom: Map with the interpretative fault system. Such faults, often called step-over faults, connect two main faults that are displaced from each other.
Tomographic images of the fault system near the Koyna and Warna reservoirs, India. Catchings’ studies showed that the main faults correspond to zones of low seismic wave velocities.
Further Resources
U.S. Quarternary Faults https://usgs.maps.arcgis.com/apps/webappviewer/index.html?id=5a6038b3a1684561a9b0aadf88412fcf